| mterry | kdub, poke about libhybris | 00:10 |
|---|---|---|
| kdub | hello mterry i did run it just now on my nexus 7, seems to work when you force the fallback mode | 00:11 |
| mterry | kdub, awesome! So no objection to a saucy upload? | 00:11 |
| kdub | well, not from me... but I'm not sure what the hybris landing requirements are exactly :) | 00:12 |
| kdub | the n7 hwc hal doesnt hybrisize well, but the rest of the driver looked ok | 00:16 |
| === _salem is now known as salem_ | ||
| === salem_ is now known as _salem | ||
| === mmrazik is now known as mmrazik|afk | ||
| === mmrazik|afk is now known as mmrazik | ||
| === alan_g is now known as alan_g|afk | ||
| === alan_g|afk is now known as alan_g | ||
| mzanetti | greyback: hey | 08:15 |
| greyback | mzanetti: hey ho | 08:16 |
| mzanetti | greyback: is your branch to be reviewed or you still fighting the notes-app? | 08:16 |
| greyback | mzanetti: still fighting | 08:16 |
| mzanetti | greyback: ok. ping me when you want it reviewed | 08:16 |
| greyback | mzanetti: will do | 08:16 |
| nic-doffay | Saviq, as soon as my sdk branch lands I'll have the orientation unity branch up for review. | 08:24 |
| nic-doffay | It's been approved so it should be shortly. | 08:24 |
| Saviq | nic-doffay, ok cool, did the performance improve when you applied those tweaks? | 08:25 |
| nic-doffay | Saviq, sadly not. | 08:25 |
| Saviq | nic-doffay, means there's probably more to be done | 08:25 |
| nic-doffay | Saviq, agreed. | 08:25 |
| nic-doffay | The review atm is a work in progress. | 08:25 |
| Saviq | nic-doffay, k | 08:26 |
| nic-doffay | If you wouldn't mind just leaving your comments then I'll get to them. | 08:26 |
| seb128 | MacSlow, hey | 08:32 |
| MacSlow | seb128, hey there | 08:37 |
| seb128 | MacSlow, hey, how are you? | 08:38 |
| seb128 | MacSlow, did you see https://code.launchpad.net/~larsu/notify-osd/fix-1189281/+merge/168550 ? can you review/approve it if you are happy with it? that fixes the segfault in saucy | 08:38 |
| MacSlow | seb128, looking at Jenkins' console output... | 08:42 |
| seb128 | MacSlow, the builds are fine, it has coverity errors but that's not new and it's complaining about the missing commit message | 08:44 |
| seb128 | MacSlow, I've tested that patch locally and confirmed it fixes the segfault if that's worth anything | 08:44 |
| MacSlow | seb128, the change itself looks ok... I just wonder if it passes "make check" | 08:45 |
| seb128 | MacSlow, it does locally here | 08:48 |
| Saviq | nic-doffay, please don't self-approve | 08:49 |
| sil2100 | larsu: ping | 08:49 |
| Saviq | nic-doffay, especially since CI failed | 08:49 |
| nic-doffay | Saviq, k | 08:50 |
| nic-doffay | Saviq, who else should I get to review the branch? | 08:52 |
| MacSlow | seb128, on my Saucy machine it does not... looking into it | 08:53 |
| seb128 | MacSlow, what is wrong? | 08:55 |
| MacSlow | seb128, make check just hangs there after the second line of "make check-TESTS" | 08:57 |
| seb128 | MacSlow, it took like 30s here and I started getting notifications | 08:57 |
| MacSlow | seb128, that doesn't happen on raring | 08:58 |
| seb128 | MacSlow, does it happen on saucy before lars' patch? | 08:58 |
| MacSlow | seb128, just trying | 08:58 |
| Saviq | mzanetti, btw, did you read michi's answer? it's a lot of what I would write | 08:59 |
| mzanetti | Saviq: half way through. got distracted but will continue reading now | 08:59 |
| Saviq | mzanetti, k | 08:59 |
| Saviq | mzanetti, I just have some things to add re: potential style exception for obviously-Qt code and some QML things that I will send to discuss on the ML | 09:00 |
| MacSlow | seb128, ah... well then... this 30 sec period happens with and without Lars' patch on saucy... still odd | 09:01 |
| MacSlow | seb128, but I'm ok with the patch | 09:02 |
| seb128 | MacSlow, thanks, can you ack it with a commit message then please? ;-) | 09:02 |
| MacSlow | seb128, what about the coverity-failure? | 09:05 |
| seb128 | MacSlow, those are not new either, we should fix it but not block the segfault fix on that imho | 09:05 |
| MacSlow | ko | 09:05 |
| MacSlow | ok | 09:05 |
| MacSlow | seb128, I'll set the status to "Approved" manually to speed things up | 09:16 |
| seb128 | MacSlow, I just did it | 09:16 |
| seb128 | MacSlow, you have to set it manually anyway, jenkins doesn't do that for you | 09:17 |
| nic-doffay | Saviq, that branch should be landing now. Bug in launchpad was causing jenkins to fail. | 09:29 |
| Saviq | nic-doffay, yes, saw it | 09:30 |
| greyback | mzanetti: hey, any tips on reproducing a segv that jenkins gets, on your local machine? In theory a raring pbuilder would be enough, no? | 09:40 |
| mzanetti | greyback: yes... probably ptest is the easiest way | 09:41 |
| greyback | mzanetti: ok thanks | 09:41 |
| mzanetti | greyback: when executing qmluitests? | 09:41 |
| greyback | mzanetti: yeah | 09:41 |
| sil2100 | bregma, dednick: hi! | 09:41 |
| greyback | mzanetti: my StageManager tests segv, apparantly before even managing to run one test. I'm not sure why | 09:42 |
| sil2100 | bregma, dednick: can you take a look at the recent unity failures in jenkins? | 09:42 |
| sil2100 | It's over the threshold right now... | 09:42 |
| === alan_g is now known as alan_g|tea | ||
| nic-doffay | Saviq, in the mean time any idea why you think the orientation is so slow? So I can work on making it quicker. | 09:43 |
| sil2100 | Trevinho: ^ | 09:46 |
| dednick | sil2100: yo | 09:46 |
| Saviq | nic-doffay, as I said, everything is being laid out again on every frame | 09:46 |
| dednick | sil2100: can you post a link? | 09:46 |
| dednick | sil2100: bregma probably wont be on for awhile | 09:47 |
| Saviq | nic-doffay, because the width/height changes | 09:47 |
| sil2100 | dednick: pasted | 09:47 |
| sil2100 | Ok, I go lie down now for sure | 09:47 |
| Saviq | nic-doffay, so we need to be smart and do some tricks | 09:48 |
| dednick | sil2100: looks pretty bawked across the board there. | 09:48 |
| Saviq | nic-doffay, I'd need a deeper look, but first of all: nothing that's not on screen (e.g. the dash when there's an app focused or the phone is locked) should animate | 09:50 |
| Saviq | nic-doffay, that's why you added the property on OrientationHelper | 09:50 |
| Saviq | nic-doffay, so that you can do some smart things while rotating | 09:50 |
| Saviq | nic-doffay, I don't have ready-made answers, though | 09:51 |
| Saviq | nic-doffay, it's probably going to be a combination of changing anchors, layering and visibility that will be the solution | 09:52 |
| Saviq | nic-doffay, for example, the panel and launcher should probably not even be inside the Helper | 09:52 |
| dednick | sil2100: dash failures seem to be caused by 'super' not showing dash. Same with hud not shoing with 'alt'. | 09:52 |
| Saviq | nic-doffay, and instead they should go offscreen and hide, wait for the rotation to finish, and come back on screen then (the launcher should stay out, too) | 09:53 |
| Saviq | nic-doffay, but we won't get there in a single go, most probably | 09:53 |
| dednick | sil2100: but have no idea if that's a random failure of keymappings or a new saucy issue. | 09:53 |
| Saviq | nic-doffay, we'll go with something initially and then expand on it later | 09:53 |
| nic-doffay | Saviq, wouldn't making it invisible accomplish the same thing as putting them off screen? | 09:55 |
| Saviq | nic-doffay, not visually | 09:55 |
| Saviq | nic-doffay, IMO the best visual experience would be to slide them out | 09:56 |
| Saviq | nic-doffay, and then make invisible | 09:56 |
| nic-doffay | Saviq, ah right, I assumed it was just for speed. | 09:56 |
| nic-doffay | Saviq, from the left? | 09:56 |
| nic-doffay | Saviq, make them invisible then slide them out after? | 09:56 |
| Saviq | nic-doffay, then rotate/re-anchor when rotating = false, make visible and slide the panel back in | 09:57 |
| Saviq | nic-doffay, if you make them invisible first then they'd just disappear in a single frame, we need an animation | 09:57 |
| nic-doffay | Saviq, yeah. So basically: Step 1 - trigger slide out animation on panel and launcher. Step 2 - rotate/re-anchor when panel and launcher have finished animation. Step 3 - trigger slide in animation on panel and launcher once rotation is done. | 09:59 |
| Saviq | nic-doffay, 2 - rotate/re-anchor when OrientationHelper.rotating becomes false | 09:59 |
| Saviq | nic-doffay, 3 - only slide the panel in, not the launcher | 09:59 |
| Saviq | nic-doffay, even if it was on screen, we don't want to bring it back straight away | 10:00 |
| nic-doffay | Saviq, ah I see you mean rotate/re-anchor the panel and the launcher, correct? | 10:00 |
| Saviq | yes | 10:00 |
| nic-doffay | gotcha | 10:00 |
| === greyback is now known as greyback|shops | ||
| nic-doffay | Saviq, I'm going to start this now. Are there existing animations I should be using? | 10:00 |
| Saviq | nic-doffay, they're not fast enough, probably | 10:01 |
| Saviq | nic-doffay, actually that made me look | 10:01 |
| Saviq | nic-doffay, OrientationHelper doesn't expose the current angle | 10:01 |
| Saviq | nic-doffay, http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/trunk/view/head:/modules/Ubuntu/Components/OrientationHelper.qml#L60 | 10:02 |
| nic-doffay | Saviq, isn't that what it's doing on line 60? | 10:02 |
| Saviq | nic-doffay, but it's not exposed | 10:02 |
| Saviq | nic-doffay, the __ | 10:02 |
| nic-doffay | Saviq, ah I see. | 10:03 |
| Saviq | nic-doffay, you could directly alk to Screen | 10:03 |
| Saviq | talk | 10:03 |
| Saviq | nic-doffay, but I think it's better to make sure you're in sync with the OrientationHelper | 10:03 |
| nic-doffay | Saviq, I agree, the int is there already, might as well use it. | 10:03 |
| nic-doffay | So I guess this needs another modification. | 10:03 |
| Saviq | nic-doffay, yeah, I un-approved https://code.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/orientation-helper-anim-alias/+merge/168489 | 10:04 |
| Saviq | nic-doffay, please include it there | 10:04 |
| nic-doffay | Saviq, cool. | 10:04 |
| === alan_g|tea is now known as alan_g | ||
| === om26er is now known as om26er|away | ||
| * greyback going out for lunch, bbiab | 11:00 | |
| dednick | Saviq: ping. | 11:07 |
| Saviq | dednick, pong | 11:07 |
| dednick | Saviq: https://jenkins.qa.ubuntu.com/job/unity-phablet-qmluitests/1236/console. Is this because haven't got a indicator-client package in debian/control ? | 11:07 |
| Saviq | dednick, in debian/*install, rather | 11:09 |
| Saviq | dednick, we need indicators-client for autopilot tests, do we? | 11:10 |
| dednick | Saviq: yes. at the moment anyway. | 11:10 |
| Saviq | dednick, we should probably package it separately | 11:10 |
| Saviq | dednick, so we don't install it (nor the /usr/share/indicators files) with unity8 | 11:11 |
| dednick | Saviq: i thought it was, apparently i'm doing it wrong though | 11:11 |
| Saviq | dednick, a completely new package, on which unity8-autopilot depends is probably best | 11:11 |
| Saviq | dednick, other than that you need to install plugins/IndicatorsClient somewhere | 11:12 |
| Saviq | dednick, although it should be "modules" if there's qml files in there | 11:12 |
| Saviq | dednick, plugin == C++ only, import == QML only, module == C++ and QML | 11:12 |
| dednick | Saviq: ic | 11:13 |
| === mmrazik is now known as mmrazik|lunch | ||
| Saviq | dednick, there's the fake-env package that we might want to use | 11:16 |
| Saviq | dednick, but also, we should probably not install anything temporary in /usr/share, we should set XDG_DATA_DIR instead to some custom prefix where we need it | 11:17 |
| dednick | Saviq: fake-env? | 11:17 |
| Saviq | dednick, there's a unity8-fake-env package | 11:17 |
| Saviq | dednick, that includes the LightDM mocks, for example, and which unity8-autopilot depends on | 11:18 |
| === MacSlow is now known as MacSlow|lunch | ||
| nic-doffay | Saviq, is it safe to move the entire overlay out the orientationHelper? | 11:20 |
| Saviq | nic-doffay, what overlay? | 11:21 |
| Saviq | nic-doffay, but anyway, yes, the OHelper is just a rectangle that rotates with a transition | 11:21 |
| Saviq | nic-doffay, so I see no reason why not | 11:21 |
| nic-doffay | Saviq, just to clarify the Item with the overlay id in Shell.qml. | 11:23 |
| nic-doffay | None of the components in there are meant to rotate? | 11:24 |
| Saviq | nic-doffay, good question | 11:24 |
| nic-doffay | just want to check if I'm missing something. | 11:24 |
| Saviq | nic-doffay, I think the only question there is the hud | 11:24 |
| Saviq | nic-doffay, but I think we should be fine with dismissing the hud when you rotate | 11:24 |
| Saviq | at least until someone tells us otherwise | 11:24 |
| nic-doffay | Saviq, ok great. | 11:24 |
| Saviq | nic-doffay, there should be no reason to animate all the Stages, too | 11:25 |
| Saviq | nic-doffay, right now the SideStage app doesn't move anyway | 11:26 |
| Saviq | nic-doffay, uh it's tricky on the tablet... | 11:27 |
| Saviq | nic-doffay, since the apps don't move, just rotate in-place | 11:27 |
| Saviq | nic-doffay, so we suddenly have a BottomStage... | 11:27 |
| Saviq | or TopStage, depending on how you look at it | 11:27 |
| Saviq | nic-doffay, and that's something I don't have an answer for atm | 11:29 |
| === s1aden is now known as sladen | ||
| nic-doffay | Saviq, I'll see what it looks like. | 11:30 |
| === mmrazik|lunch is now known as mmrazik | ||
| === mzanetti is now known as mzanetti|lunch | ||
| nic-doffay | Saviq, I've moved what I currently can out the Orientation Helper. I'm going to start working on what we discussed earlier then we can have a further look at what more could be done. | 11:54 |
| nic-doffay | Sound reasonable? | 11:54 |
| Saviq | nic-doffay, sure | 11:54 |
| dednick | Saviq: so with the IndicatorsClient module, does that need to be in a separate source folder then? ie. not in plugins, or is it just the install folder that should be modules/IndicatorsClient? | 12:08 |
| Saviq | dednick, separate source, please | 12:09 |
| dednick | Saviq: ok | 12:09 |
| Trevinho | Cimi: I think I finally found what was wrong in libbamf... | 12:15 |
| Trevinho | Cimi: can you give me the bug #? | 12:16 |
| === MacSlow|lunch is now known as MacSlow | ||
| Saviq | paulliu, what's the difference between (< 7.80) and (<< 7.80)? | 12:32 |
| paulliu | Saviq: < is obsolete. And < means <= IIRC. Better to only use <= or << | 12:33 |
| Saviq | paulliu, good to know, thanks | 12:33 |
| larsu | sil2100: hey | 12:41 |
| === mzanetti|lunch is now known as mzanetti | ||
| dandrader | mzanetti, ping | 12:46 |
| mzanetti | hi dandrader | 12:47 |
| dandrader | hi | 12:47 |
| dandrader | would you have time to review this? https://code.launchpad.net/~dandrader/unity/8_directionalDragMinSpeed/+merge/168136 | 12:47 |
| === _salem is now known as salem_ | ||
| mzanetti | dandrader: ok. doing that now | 12:47 |
| dandrader | mzanetti, thanks! | 12:54 |
| === alan_g is now known as alan_g|lunch | ||
| === boiko_ is now known as boiko | ||
| === jhodapp is now known as jhodapp|afk | ||
| Saviq | mzanetti, standup | 13:32 |
| Saviq | dednick, nic-doffay, standup | 13:33 |
| Cimi | lp:ubuntu-settings-components dednick | 13:37 |
| greyback | sorry, something gone wrong on my machine, trying to re-join | 13:38 |
| mzanetti | Cimi: btw... was my review on the calendar tests helpful? Or do you want me to explain something more in depth? | 13:43 |
| Cimi | ohyeah | 13:44 |
| Cimi | mzanetti, I have a question for you on the first test | 13:44 |
| mzanetti | Cimi: hit me | 13:44 |
| Cimi | mzanetti, the others got changes | 13:44 |
| Cimi | mzanetti, I don't know what to check for the compressed state | 13:44 |
| Cimi | mzanetti, the dimension isn't really 6 times | 13:44 |
| Cimi | there's padding etc | 13:44 |
| mzanetti | Cimi: ah, I see... | 13:45 |
| mzanetti | Cimi: is it too complicated to quickly sum up the padding in the test code? | 13:45 |
| Cimi | ? | 13:45 |
| mzanetti | Cimi: something like compare(newHeight, "oldHeight * 6 + spacing * 5") | 13:45 |
| Cimi | ah | 13:46 |
| Cimi | old height contains padding | 13:46 |
| Cimi | so it's not that immediate | 13:46 |
| mzanetti | Cimi: sure... I don't have the code here, but you get the idea... | 13:46 |
| Cimi | unless you change code just for that | 13:46 |
| mzanetti | Cimi: if thats not feasible, I'd probably go for something like this: | 13:47 |
| Cimi | which is then useless imho, at this point you don't test at all | 13:47 |
| mzanetti | Cimi: verify(newHeight > oldHeight * 5) | 13:47 |
| mzanetti | Cimi: so the test verifies that it actually expads... if there are some elements in the expanding area that change visible to true while expanded you could check that too in the test | 13:48 |
| Cimi | ah ok | 13:48 |
| Cimi | will have it a go | 13:48 |
| mzanetti | cool | 13:48 |
| === alan_g|lunch is now known as alan_g | ||
| mzanetti | whats the minimum battery level to flash the phone? | 13:49 |
| sil2100 | larsu: heya | 13:50 |
| sil2100 | larsu: so, we have some HUD bugs that are blocking releases, and tedg was working on those yesterday - he put up 2 merges that need reviewing | 13:51 |
| sil2100 | larsu: did you take a look at those already? | 13:51 |
| dednick | Cimi: have you integrated those components into any of the indicators yet? | 13:53 |
| larsu | sil2100: no, I don't usually review hud stuff - I'm not very familiar with the code | 13:53 |
| Cimi | dednick, nope | 13:53 |
| Cimi | dednick, it's not my task | 13:53 |
| Cimi | at the moment | 13:53 |
| Cimi | now I am looking into theming | 13:53 |
| dednick | Cimi: sure. | 13:53 |
| larsu | sil2100: is there noone else that could look at them? | 13:53 |
| dednick | Cimi: what you need me to review? pretty much everything there? | 13:55 |
| Cimi | dednick, yeah | 13:55 |
| Cimi | dednick, it's more an easy review, like having a look and tell me if you don't like things | 13:55 |
| sil2100 | larsu: hmm, not sure, let me ask pete-woods if he could take a look | 13:55 |
| sil2100 | pete-woods: ping | 13:55 |
| dednick | Cimi: ok. you had the design guys go over it yet? or is that after me? | 13:56 |
| Cimi | nope | 13:56 |
| sil2100 | dednick: ping | 14:10 |
| dandrader | Saviq, any estimates as to when unity8 will work normally (with all dashes) in saucy? | 14:11 |
| Saviq | dandrader, https://code.launchpad.net/~unity-team/unity/8.new-libunity/+merge/167733 | 14:11 |
| Saviq | dandrader, so, this or early next week | 14:11 |
| sil2100 | dednick: so, regarding those unity failures - is that the case for only one test-machine? | 14:12 |
| sil2100 | dednick: or was that happening on both? | 14:12 |
| dednick | sil2100: looks like it was just the intel machine | 14:14 |
| dednick | not sure what was happening on the ati box | 14:14 |
| sil2100 | Still a bit too much, since it's 19 failures on ati | 14:18 |
| sil2100 | Intel had 22 | 14:18 |
| sil2100 | dednick: from what I see, unity seems a bit broken | 14:19 |
| sil2100 | Trevinho, andyrock: ping | 14:19 |
| Trevinho | sil2100: pong | 14:19 |
| sil2100 | Are you very busy? ;) | 14:20 |
| sil2100 | Trevinho, andyrock: since I would need some volunteers to join a quest of regression hunting | 14:20 |
| pete-woods | larsu: hi, what's the change you want looking at? | 14:22 |
| larsu | pete-woods: I don't know... sil2100 asked me :) | 14:22 |
| pete-woods | oh, okay! | 14:23 |
| pete-woods | ping sil2100! | 14:23 |
| Trevinho | sil2100: mh, then? | 14:23 |
| sil2100 | pete-woods: hi! | 14:24 |
| pete-woods | sil2100: I understand you want a MR looking at or something like that? | 14:24 |
| sil2100 | Trevinho: since the latest unity test run returned many failures, some seem to be related to hud issues, some to keybindings... | 14:24 |
| sil2100 | pete-woods: yes ;) Since I think tedg made some hud fixes | 14:25 |
| dandrader | mterry, a leftover from yesterday's merge proposal: https://code.launchpad.net/~dandrader/unity/8_stageDDACleanup/+merge/168676 | 14:25 |
| sil2100 | pete-woods: I remember you were familiar with the code, yes? | 14:25 |
| pete-woods | sil2100: somewhat! | 14:25 |
| didrocks | fginther: around? | 14:25 |
| sil2100 | pete-woods: I see that one of the merges got reviewed and approved already, but the other one: | 14:25 |
| sil2100 | https://code.launchpad.net/~ted/hud/proper-client-names/+merge/168518 | 14:25 |
| fginther | didrocks, morning | 14:25 |
| didrocks | fginther: hey! | 14:25 |
| didrocks | fginther: do you have a minute for https://code.launchpad.net/~didrocks/cupstream2distro-config/python-upa-rename/+merge/168677? | 14:26 |
| fginther | didrocks, certainly | 14:26 |
| didrocks | fginther: I think you need to redeploy something on your side as well :) | 14:26 |
| tedg | pete-woods, Wellark is looking at some, but there's a few. | 14:26 |
| sil2100 | tedg: hi! Do the two merges fix the issues with 'no results returned' I poked you about yesterday? | 14:26 |
| didrocks | thanks! | 14:26 |
| pete-woods | sil2100: looks like Wellark beat me to that one | 14:26 |
| tedg | sil2100, The bamf-focus-fix one was the core issue, the others were things I found along the way. | 14:26 |
| tedg | Stupid BAMF changes. | 14:27 |
| tedg | :-) | 14:27 |
| * tedg is really excited to drop dbus-glib there | 14:27 | |
| tedg | Should be *much* fewer BAMF issues in the future. | 14:27 |
| sil2100 | tedg: \o/ so, once the bamf fix gets merged, you think it's safe to rebuild the hud stack and check the test results again? | 14:28 |
| fginther | didrocks, thanks for the heads up, I'll deploy on our side when the merge completes | 14:30 |
| didrocks | thanks fginther :) | 14:30 |
| tedg | sil2100, Yeah, since there's a few queued to land I'd wait for all of them though :-) | 14:31 |
| * didrocks is happy that tests finds real regressions :) | 14:31 | |
| mzanetti | dandrader: hey, reading through the diff the right edge (Stage) should already use that minSpeed thing, right? | 14:33 |
| dandrader | mzanetti, yes | 14:34 |
| sil2100 | didrocks: those fixes should also resolve some unity problems, although there's still one issue in unity that just irritates me | 14:34 |
| mzanetti | dandrader: on the device, if I press the right edge without moving, it behaves weird | 14:34 |
| mzanetti | dandrader: it starts moving the app, then it releases the app and grabs it again, resulting in weird flickering | 14:35 |
| dandrader | mzanetti, I'll check | 14:35 |
| didrocks | sil2100: ok, good hunt! do we have stuff that we should publish (apart from QA where I'm running the build with the renamed component) | 14:36 |
| didrocks | ? | 14:36 |
| === jhodapp|afk is now known as jhodapp | ||
| sil2100 | didrocks: are we still blocked with apps on the copyright issues? | 14:39 |
| didrocks | sil2100: no, everything's merged (same for mediaplayer in the media stack) | 14:41 |
| didrocks | sil2100: I removed this morning the "force manual publishing" | 14:41 |
| sil2100 | didrocks: ok, so I think we can publish both apps and media, since they're passing nicely anyway - there have only been AP-related issues with those | 14:42 |
| mzanetti | mterry: btw. I replied to your comments on the MR regarding the lockscreens | 14:42 |
| didrocks | sil2100: so the stack deps on hud/indicators are not blocking and the packaging changes are good? | 14:42 |
| sil2100 | didrocks: indicators, well, I need to check the vids again, since first it was an jenkins issue and now just some u-g-m things failed | 14:42 |
| didrocks | sil2100: maybe you want me to have a second look on the packaging changes? | 14:42 |
| mterry | mzanetti, yup, will look again | 14:42 |
| sil2100 | didrocks: I'll re-check the packages, one moment | 14:43 |
| mzanetti | mterry: I don't really see why I should make it complicated and copy/paste the whole test suite just to use the single user backend it the full backend would fit exactly my needs | 14:43 |
| mzanetti | s/it/if/ | 14:43 |
| mterry | mzanetti, well... I guess I didn't see why there was a has-pin user in full/ anyway. Since you already have single-full/. Not a problem per se, just didn't seem necessary, especially since for tablet testing with full/, we wouldn't be showing a PIN pad anyway. I don't know what you mean by copy/pasting a whole test suite? | 14:46 |
| sil2100 | didrocks: I only checked gallery-app now, but I already see a small copyright-packaging problem already | 14:46 |
| didrocks | oh? | 14:46 |
| sil2100 | didrocks: the .cmake files in cmake/, some have a different license than GPL-3 | 14:46 |
| sil2100 | didrocks: as they're copyrighted by cmake guys and GPL-2 | 14:47 |
| mterry | mzanetti, ah I see. You mean for the passphrase vs pin | 14:47 |
| sil2100 | More of a nit-pick, but still | 14:47 |
| didrocks | sil2100: we don't really care about build systems | 14:47 |
| mzanetti | mterry: I run all the test functions twice (by using _data() functions) | 14:47 |
| didrocks | sil2100: see my email from this morning on the new components :) | 14:47 |
| mzanetti | mterry: and if I would need to run it with different LD_LIBRARY_PATHs I would need to rip it apart | 14:47 |
| mzanetti | copy/pasting the test functions | 14:47 |
| * sil2100 now read that | 14:48 | |
| sil2100 | Oh, ok | 14:48 |
| mterry | mzanetti, so no tests that cover the Shell.qml bit? It would be good to have an autopilot test maybe? | 14:48 |
| mzanetti | mterry: definitely an autopilot thing, however, I don't think that code will stay in shell.qml for long, does it? | 14:49 |
| sil2100 | Since Mirv was pointing out the cmake/ files so I was fixing those as asked | 14:49 |
| sil2100 | didrocks: nevermind that then ;) | 14:49 |
| mterry | mzanetti, it will move to something similar, a GreeterShell.qml or whatever | 14:49 |
| didrocks | :) | 14:49 |
| didrocks | ok, gallery-app sounds good to me | 14:49 |
| mzanetti | mterry: I would suggest once we move the stuff away from unity8 into a real lightdm greeter we get together and write autopilot integration tests for that setup together | 14:50 |
| didrocks | sil2100: so you can publish the app stack if there is no issue with the stack dep (just tell me so that I can NEW the apps right away) | 14:50 |
| mzanetti | mterry: writing tests with all the shell objectNames etc would need to be replaced anyways in the next step | 14:50 |
| mterry | mzanetti, it's not so bad. https://code.launchpad.net/~mterry/unity/8.split/+merge/168688 keeps most of the greeter tests intact | 14:51 |
| mterry | mzanetti, the journey for autopilot won't change, we'll just launch a different executable before testing a given test | 14:52 |
| sil2100 | didrocks: ok, deps look good, I'll publish | 14:52 |
| mzanetti | mterry: sure, but its not in Shell.qml any more? | 14:52 |
| sil2100 | didrocks: will re-check media and publish as well | 14:52 |
| mterry | mzanetti, the greeter stuff is pulled into a separate GreeterShell.qml, at least in my branch | 14:53 |
| didrocks | sil2100: hum, let me recheck the packaging diff (normally only people with upload rights should ack, but we'll get that fixed for you and Mirv by the end of cycle ;)) | 14:53 |
| sil2100 | didrocks: I'm still a bit inexperienced in that, some things I miss, but slowly getting a hang of it | 14:54 |
| didrocks | sil2100: it's just that packaging diff should be check with people with upload rights (it's the contract with the release team) | 14:55 |
| sil2100 | didrocks: give me a sign if it's green and I publish then ;) | 14:56 |
| didrocks | sil2100: just checked it, the packaging diff looks good to me :) | 14:56 |
| didrocks | (but feel free to look at it as well and raise questions if any ;)) | 14:56 |
| sil2100 | Sickness calls, brb! | 14:56 |
| nic-doffay | Saviq, at the moment the OrientationHelper's states are called whenever the rotation angle changes. This is a core part of the OrientationHelper. Any ideas on how I should get these state changes to trigger after the launcher and the panel, both of which are located in Shell.qml? | 15:00 |
| sil2100 | Ok, publishing! | 15:00 |
| nic-doffay | I mean the animating bit is easy, but there's a lot of code here that has been done without all of this in mind. | 15:00 |
| mmrazik | didrocks: quick question regarding adding stuff to daily release -- I'm going to add https://launchpad.net/libusermetrics to the indicators stack (with daily_release: False). Is there any process to make sure you notice new projects and can (eventually) daily_release: True-them? | 15:00 |
| Saviq | nic-doffay, you mean to delay the OrientationHelper's rotation? | 15:02 |
| nic-doffay | Saviq, yeah | 15:03 |
| didrocks | mmrazik: no process, but this week, I'll get everyone looking at what's in the config with daily_release: False | 15:03 |
| didrocks | mmrazik: for the future, pings are enough :) | 15:03 |
| mmrazik | didrocks: ok | 15:03 |
| sil2100 | \o/ | 15:03 |
| didrocks | thanks mmrazik :) | 15:03 |
| nic-doffay | Saviq, it's tricky because it's a part of the SDK and technically it shouldn't be delayed. | 15:03 |
| nic-doffay | imo at least | 15:03 |
| Saviq | nic-doffay, it could have a configurable delay, though | 15:05 |
| === alan_g is now known as alan_g|tea | ||
| Saviq | nic-doffay, 0 by default | 15:05 |
| nic-doffay | Saviq, that's true. How do you delay states though? | 15:05 |
| nic-doffay | At the moment whenever the orientationAngle changes it toStrings to the state. | 15:05 |
| Saviq | nic-doffay, you don't, you add a PauseAnimation in the transition | 15:06 |
| Saviq | nic-doffay, but | 15:06 |
| sil2100 | didrocks: apps and media stacks are being published, noticed a failure in the QA stack - should I restart it? Is it because of the renaming? (I see it's missing python-ubuntu-platform-api in the prepare step) | 15:06 |
| Saviq | nic-doffay, delaying would mean that the app's rotation gets out of sync with the shell rotation | 15:06 |
| nic-doffay | Saviq, exactly. | 15:06 |
| Saviq | nic-doffay, but then if we're talking that the panel should go away anyway | 15:06 |
| nic-doffay | Saviq, I'm mainly worried about the panel going away at the correct time. | 15:07 |
| nic-doffay | Not while the rotation is happening. | 15:07 |
| nic-doffay | Which is the case atm. | 15:07 |
| didrocks | sil2100: oh weird, I redeployed with that | 15:09 |
| didrocks | /var/log/upstart/otto-setup.log: E: Package 'python-ubuntu-platform-api' has no installation candidate | 15:11 |
| didrocks | sil2100: seems different? | 15:11 |
| didrocks | ah, it's armhf only | 15:11 |
| didrocks | sil2100: I guess for now, I need to remove it | 15:12 |
| didrocks | from the packages to install | 15:12 |
| dandrader | Saviq, is run_on_device working fine for you on a freshly-flashed device(with phablet-flash and no /home/phabler/shell dir)? | 15:12 |
| Saviq | dandrader, checking | 15:13 |
| === alan_g|tea is now known as alan_g | ||
| didrocks | sil2100: not sure if autopilot-touch is armhf only. If not, that can block the migration to -proposed | 15:14 |
| didrocks | sil2100: let's keep an eye on it | 15:14 |
| dandrader | Saviq, I get this http://paste.ubuntu.com/5755165/ | 15:14 |
| sil2100 | didrocks: ah, hm, indeed | 15:14 |
| didrocks | restarted with "foo" | 15:14 |
| Saviq | dandrader, never saw that | 15:16 |
| dandrader | :( | 15:16 |
| dandrader | Saviq, so all worked fine for you now? | 15:16 |
| Saviq | dandrader, just checking, didn't flash yet, just dropped "shell" | 15:16 |
| Saviq | dandrader, but it went past that error for me | 15:17 |
| Saviq | dandrader, /me flashes | 15:17 |
| Saviq | nic-doffay, problem is that even if you delay the OHelper in the shell, the app won't know about it | 15:19 |
| Saviq | nic-doffay, so it will rotate regardless | 15:19 |
| nic-doffay | Saviq, yeah. I'm not sure how to get around all of this. | 15:19 |
| Saviq | nic-doffay, no way around it now, we can't talk to the app about this yet | 15:19 |
| dandrader | Saviq, where that build.ninja comes from? | 15:20 |
| nic-doffay | Saviq, right for the time being I'll just make the hide animation very quick. | 15:20 |
| Saviq | dandrader, ninja is the build system we switched to | 15:20 |
| Saviq | nic-doffay, yup, SnapDuration | 15:20 |
| nic-doffay | Also, who is an expert on the panel? | 15:20 |
| Saviq | nic-doffay, dednick | 15:20 |
| nic-doffay | Saviq, where is SnapDuration? | 15:20 |
| dandrader | Saviq, I mean that file the error is complaining about | 15:20 |
| Saviq | nic-doffay, UbuntuAnimation.SnapDuration | 15:21 |
| nic-doffay | Saviq, cool ta | 15:21 |
| Saviq | nic-doffay, from Ubuntu.Components | 15:21 |
| Saviq | dandrader, it should create it... | 15:21 |
| Saviq | dandrader, cmake -G Ninja should | 15:21 |
| nic-doffay | Saviq, one more thing. I get loads of reference errors when using UbuntuAnimation. | 15:22 |
| dandrader | Saviq, ah, that's likely the problem. cmake is not ther. run_on_device --setup doesn't install it | 15:22 |
| nic-doffay | It's probably because it's C++ | 15:22 |
| Saviq | dandrader, it does, but must've failed somehow | 15:22 |
| Saviq | nic-doffay, what do you mean? | 15:23 |
| Saviq | nic-doffay, it's there, doesn't matter if it's C++ or not | 15:23 |
| nic-doffay | Saviq, with UbuntuComponents included it gives me a reference error. | 15:23 |
| nic-doffay | Whenever I UbuntuAnimation. anything. | 15:23 |
| nic-doffay | It works however. | 15:23 |
| nic-doffay | which is perplexing | 15:23 |
| Saviq | nic-doffay, can you paste the exact error? | 15:23 |
| Saviq | nic-doffay, and your toolkit version? | 15:24 |
| nic-doffay | Saviq, https://pastebin.canonical.com/92548/ | 15:24 |
| dandrader | Saviq, got it: "E: Unable to find a source package for unity8" | 15:24 |
| dandrader | from run_on_device --setup | 15:24 |
| nic-doffay | Saviq, where can I find the toolkit version? | 15:24 |
| Saviq | nic-doffay, apt-cache policy qtdeclarative5-ubuntu-ui-toolkit-plugin | 15:25 |
| Saviq | dandrader, the most recent run_on_device should use mk-build-deps | 15:25 |
| Saviq | dandrader, which just builds a package depending on the build deps from debian/control | 15:25 |
| dandrader | Saviq, hmm, ok. gonna rebase my branch then | 15:25 |
| nic-doffay | Saviq, 0.1.46daily13.06.05.1bzr538saucy0 | 15:25 |
| Saviq | nic-doffay, yeah, the newest one | 15:26 |
| nic-doffay | Saviq, I'll ignore it for now. | 15:26 |
| Saviq | nic-doffay, `qmlplugindump Ubuntu.Components 0.1 | grep -i animation` ? | 15:27 |
| Saviq | nic-doffay, and ignore the font complaints | 15:27 |
| Saviq | dandrader, so yeah, rebasing will help | 15:27 |
| nic-doffay | Saviq, looks ok to me: https://pastebin.canonical.com/92551/ | 15:28 |
| Saviq | nic-doffay, yeah | 15:28 |
| Saviq | nic-doffay, not sure what's happening, we'll see over the review | 15:28 |
| dandrader | Saviq, yeah. that did it. sorry for the noise | 15:29 |
| Saviq | dandrader, cheers | 15:29 |
| nic-doffay | Saviq, agreed. Should have some additional stuff for commentary soon too. | 15:29 |
| Saviq | dandrader, it should be much more robust no | 15:29 |
| Saviq | w | 15:29 |
| nic-doffay | Saviq, I think there's already been a bit of a speed up moving some components out the OH | 15:29 |
| Saviq | nic-doffay, yeah, there should be | 15:30 |
| dandrader | mzanetti, when you tried out my MP, was it merged with trunk of did you try out the branch directly? | 15:30 |
| mzanetti | dandrader: directly your branch | 15:30 |
| dandrader | mzanetti, ok | 15:31 |
| Saviq | dandrader, mzanetti I think to be consistent, QML uses "velocity" usually | 15:32 |
| Saviq | dandrader, mzanetti, and it's in px/s | 15:32 |
| Saviq | or whatever the coordinate system is | 15:33 |
| Saviq | mzanetti, dandrader e.g. http://qt-project.org/doc/qt-5.0/qtquick/qml-qtquick2-flickable.html#maximumFlickVelocity-prop | 15:33 |
| dandrader | Saviq, hmm, so I can safely say that "spacial unit used by QML coordinate system" === "pixels" | 15:34 |
| Saviq | dandrader, yes | 15:34 |
| Saviq | dandrader, it's a good enough assumption | 15:35 |
| nic-doffay | Saviq, just to double check, the panel itself needs to hide, not just the dropdown on rotation? | 15:42 |
| Saviq | nic-doffay, yes, the whole panel | 15:43 |
| dednick | Saviq: why would the panel hide? it's always visible. | 15:43 |
| Saviq | dednick, only during rotation | 15:43 |
| dednick | ah | 15:43 |
| Saviq | dednick, so that we don't have to try and sync with the app | 15:43 |
| mzanetti | mterry: added autopilot tests for lockscreens | 15:46 |
| * mterry hugs mzanetti | 15:46 | |
| mterry | mzanetti, hopefully we can get this thing in today | 15:47 |
| mzanetti | mterry: I created a new testfile just for lockscreen tests and created 2 szenarios: one for pin, one for passphrase. just testing on phone form factor | 15:47 |
| mterry | mzanetti, why was "self.assertThat(pinentryField.text, Eventually(Equals("432")))" necessary? It's not done in the success mode | 15:50 |
| mzanetti | mterry: I just want to make sure that it actually types, and upon recognition of the wrong password it clears the field again | 15:55 |
| mzanetti | mterry: without that line there would be the theoretical chance that it's not even typing and hence the test still passing | 15:55 |
| mterry | mzanetti, ah... fair I guess | 15:55 |
| mterry | mzanetti, approved! Thanks man | 15:56 |
| mzanetti | mterry: cool :) | 15:56 |
| tedg | sil2100, FYI, those branches landed. | 15:58 |
| tedg | didrocks, Could we have, on hud failures (as an example), just call "ubuntu-bug hud" or something like that. So we'd get a bug with all the log files. | 15:59 |
| tedg | didrocks, Then we could use the apport hooks for that stuff. | 15:59 |
| cyphermox | sil2100: didrocks: does python-ubuntu-platform-api really belong in the QA stack? | 16:00 |
| didrocks | cyphermox: yeah, it's done and conceived for being used only for autopilot | 16:00 |
| cyphermox | ok | 16:01 |
| didrocks | tedg: define the "failure" part? :) | 16:01 |
| didrocks | tedg: you mean, autopilot tests failing? | 16:01 |
| tedg | didrocks, Yeah | 16:01 |
| didrocks | (if so, it's more an autopilot option I guess) | 16:01 |
| tedg | didrocks, So who do I harass? thomi? :-) | 16:03 |
| didrocks | tedg: yeah, or veebers | 16:03 |
| didrocks | tedg: I would recommend both :p | 16:04 |
| * tedg does jujitsu and gets veebers to harass thomi | 16:04 | |
| nic-doffay | Saviq, one more glaring issue. The panel leaves a gaping space that's quite visible upon rotation. Any ideas on what I should do to get around this? | 16:08 |
| mzanetti | dandrader: I rebased your branch on trunk and its working like a charm now | 16:10 |
| mzanetti | dandrader: and wow! feels perfect! great job | 16:10 |
| Saviq | nic-doffay, I don't think we can do anything about this, the app will fill that space anyway, when we implement that in qtubuntu | 16:11 |
| nic-doffay | Saviq, ok. There's one thing left now. The launcher and the panel still have to rotate. | 16:12 |
| nic-doffay | Obviously this isn't happening since I moved them out the OH | 16:12 |
| nic-doffay | It's a similar issue to what we spoke about earlier. Since everything happens depending on the rotating variable. | 16:12 |
| Saviq | nic-doffay, yeah, you need to have states for them that change their anchors and rotation | 16:12 |
| nic-doffay | Set the states if they aren't rotating Saviq? | 16:13 |
| Saviq | nic-doffay, yes, after rotating changes back to false, change the anchors and rotation accordingly, without transitions | 16:13 |
| Saviq | nic-doffay, and slide back in | 16:13 |
| seb128 | mhr3, not on #ubuntu-touch? | 16:14 |
| nic-doffay | Saviq, should these states live in Shell.qml or the individual Launcher/Panel.qml ? | 16:14 |
| Saviq | nic-doffay, Shell, as they don't know about the rotation, really (and shouldn't) | 16:14 |
| mhr3 | seb128, another channel i need to monitor :) | 16:15 |
| seb128 | mhr3, yeah ;-) | 16:15 |
| === dandrader_ is now known as dandrader|lunch | ||
| === om26er_ is now known as om26er | ||
| mzanetti | dandrader|lunch: ok, I'm through with the review. left 2 more comments. rest looks good | 16:34 |
| Cantide | where can one find information on the state of unity 8 for the desktop? or is it not yet at a stage which is presentable to the public? | 16:41 |
| sil2100 | tedg: hmmm, there still seem to be some issues with HUD | 17:00 |
| sil2100 | But different ones it seems | 17:00 |
| sil2100 | tedg: I need to go, but could you take a look in your free time? I'll paste the link | 17:02 |
| tedg | Free time? :-) | 17:02 |
| sil2100 | ...;) | 17:02 |
| sil2100 | Ok ok, that was a bad joke | 17:02 |
| sil2100 | Sorry about that :< | 17:02 |
| sil2100 | ;p | 17:02 |
| tedg | I think these are Trevinho's bugs. | 17:03 |
| tedg | They're all BAMF warnings. | 17:04 |
| tedg | unity <unknown>:0 Unable to fetch children: No such interface 'org.ayatana.bamf.view' on object at path /org/ayatana/bamf/application62801462 | 17:04 |
| Trevinho | tedg: yep... Not really a bug actually.. .It can happen when apps are quickly opened/closed... Also if I'm looking to avoid this as well | 17:04 |
| === mmrazik is now known as mmrazik|afk | ||
| === alan_g is now known as alan_g|life | ||
| tedg | Trevinho, Would it happen if we ask for the view before the app is realized? Guessing something like that is happening in the autopilot testing. | 17:06 |
| Trevinho | tedg: mh it could be but it's unlikely I think... | 17:08 |
| Trevinho | tedg: generally that happens when the app is closed and we try to get something that is not really on bus | 17:09 |
| tedg | Trevinho, Seems like it should notify that it is removed from the bus, no? Then libbamf returns a critical? | 17:11 |
| Trevinho | tedg: yes, that happens, but it seems there are some problems on the library that prevents to check this locally... I'll look into that soon, but it shouldn't change the result of that call btw | 17:12 |
| tedg | Trevinho, I think the only time we're doing that is on startup in HUD. | 17:13 |
| tedg | Trevinho, Basically getting the list of applications, and then the list of windows. | 17:13 |
| tedg | Trevinho, So that's why I was thinking it was a race at startup. | 17:13 |
| Trevinho | tedg: couldn't be that an app that you cached is actually closed? | 17:14 |
| tedg | Trevinho, It could be, but unlikely in the autopilot test because there's only one app. | 17:14 |
| mhr3 | Saviq, ping? | 17:16 |
| mhr3 | Saviq, sorry for the late pings recently... | 17:16 |
| mhr3 | Saviq, any recommended reading for tomorrows meeting? | 17:16 |
| mhr3 | Saviq, as in a doc that lists all the objects and their properties available to QML? | 17:17 |
| === mmrazik|afk is now known as mmrazik | ||
| === dandrader|lunch is now known as dandrader | ||
| olli_ | bregma, thanks for the update on the 13.10 BP | 18:06 |
| === olli_ is now known as olli | ||
| * bregma is whittlin' down his to-do list | 18:09 | |
| === AlanChicken is now known as AlanBell | ||
| === mmrazik is now known as mmrazik|afk | ||
| olli | bregma, ping | 19:21 |
| bregma | olli, pong | 19:21 |
| === dandrader is now known as dandrader|afk | ||
| olli | bregma, sorry... | 20:08 |
| olli | I was chatting with pmcgowan about u8 on 13.10 | 20:08 |
| olli | he suggested to add movie player & the music player app to the list | 20:09 |
| olli | and was anticipating some issues around files vs content picking etc | 20:09 |
| olli | mind adding the first to the BP (my write rights are gone:) | 20:09 |
| olli | and for the 2nd (files)... just wanted to mention it, wasn't sure if this was on your radar | 20:10 |
| === dandrader|afk is now known as dandrader | ||
| bregma | olli, done | 20:24 |
| olli | bregma, thx! | 20:24 |
| Esokrates | hi, are there already plans for dynamic workspace support in unity-next? how concrete are potential plans? | 20:24 |
| Esokrates | is design input by the community welcome, or is the design restricted (and there are already exact plans)? | 20:26 |
| === racarr_ is now known as racarr | ||
| === naee is now known as eean | ||
| === salem_ is now known as _salem | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!